The determinants and consequences of excessive oral drug self-administration by schedule-induced polydipsia will be explored, emphasizing oral cocaine (COC) and benzodiazepine (BZ) abuse variables and fine-motor performance consequences: (A: COC) (1) The amelioration of excessive COC intake by acute and chronic treatment with possible therapeutic agents (e.g., desipramine, mazindol) and by (2) alternatives to COC ingestion (e.g., other solutions, activity wheel). (3) Exacerbation of COC abuse by injections of nicotine (NIC) and caffeine (CAF). (4) The gateways to exacerbated COC abuse by first instituting a history of licit-drug polydipsia (NIC, CAF), and (5) possible behavioral immunization (history of situational alternatives) against COC abuse acquisition. The consequences of acute and chronic injection and withdrawal of COC for discriminative motor performance, as well as COC interactions with NIC and CAF. (B: BZ) (1) The amelioration of excessive BZ intake by blockers (Ro 15-1788, CGS 8216) and (2) its exacerbation by CAF or a history of sedative (ethanol) abuse. (3) Acute and chronic drug anxiolytic action evaluation by a method employing the exaggerated ingestion of NaC1 solutions by such agents. (3) The fine motor control consequences of daily rebound from chronic exposure to midazolam (e.g., as perhaps exaggerated by inverse agonist [FG 7142] probes) and of midazolam-CAF interactions. For both COC and BZs, serum levels of drug and metabolites will be measured, both with respect to chronic amounts abused and the levels associated with motor performance disruption.